This project aims to understand mixing and transport patterns of substrates (e.g. chemicals) in reactors by using the latest 3D flow measurement techniques and concepts from dynamical systems and network theory.
The identification of mixing patterns and heterogeneities in chemical and biochemical reactors is crucial for assessing their internal mixing and segregation states. Whenever the timescales of mixing are similar to those of the process of interest, e.g. the timescales of a chemical reaction, neither the simplification of perfectly mixed educts nor that of complete segregation can be made, and yield and selectivity of the process can be influenced by fluid dynamics. Lagrangian trajectories form the basis for the identification of the mixing patterns governing the fluid dynamic transport and mixing. In experimental settings these can either be derived by particle tracking velocimetry or by Lagrangian sensors that record their own track. Both data sets are highly sparse, imperfect, and affected by measurement inaccuracies. Moreover, due to inertia, sensor tracks do not exactly represent the underlying motion of the fluid and substrates (e.g. chemicals, microorganisms). The aim of this project is to identify Lagrangian transport and mixing patterns of substrates by applying novel data-based mathematical Lagrangian analysis methods on experimentally measured tracer data and sensor trajectories.
The project has therefore two main targets:
First, we aim to measure the time-resolved Lagrangian trajectories in three spatial dimensions either using time resolved particle tracking velocimetry (4D-PTV) or, later, also thespatio-temporal information of Lagrangian sensors.
Second, we aim to develop new mathematical tools tounravel the spatio-temporally resolved mixing patterns on the basis of this imperfect data that allows us to quantify the transient mixing processes in the reactor. The scales of the measurement volume depend on the reactor studied and vary from the millimetre range at internals to the meter range when large scale reactors are investigated with Lagrangian sensors. The core scientific questions of this proposal are:
- How can moving coherent compartments or dead zones be identified computationally from Lagrangian trajectories and even from sparse or imperfect experimental data by new Lagrangian mixing measures?
- How do these spatio-temporally resolved Lagrangian mixing measures relate to common mixing analytics such as concentration fields, mixing times, segregation, and mixing quality?
- How can the derived mixing measures aid to determine optimal process conditions and reactor geometry concerning the volume flow, placing of built-in sensors, components with smart surfaces or gas evolving electrodes on inlets?
Project related links:
